Automatic Thoughts Questionnaire (ATQ-N)

Abstract

The Automatic Thoughts Questionnaire (ATQ-N; frequently designated as the ATQ-30) is an established, 30-item psychometric self-report instrument engineered to measure the frequency and degree of belief associated with negative automatic thoughts (NATs)—the transient, repetitive, and unsolicited depressotypic cognitions characteristic of clinical depression and affective dysregulation. Originally operationalized by clinical psychologists Steven D. Hollon and Philip C. Kendall in 1980, the scale was constructed as an empirical operationalization of Aaron T. Beck’s cognitive model of psychopathology, which posits that systematic negative cognitive biases regarding the self, the ongoing world, and the personal future constitute the proximal cognitive diathesis for unipolar depressive episodes.

Respondents evaluate the 30 statements on a 5-point Likert frequency dimension ranging from 1 (Not at all / Never) to 5 (All the time / Always) over the preceding seven-day timeframe, yielding a total composite score bounded between 30 and 150 points. A secondary, clinically sensitive scoring metric permits individuals to rate their contemporaneous subjective “Degree of Belief” across the same items on an anchored 5-point scale from 1 (Not at all) to 5 (Totally). Exploratory and confirmatory factor analyses consistently substantiate a robust multidimensional architecture comprising four interrelated first-order latent factors: Personal Maladjustment and Desire for Change (PMDC), Negative Self-Concepts and Negative Expectations (NSNE), Low Self-Esteem (LSE), and Helplessness.

Psychometrically, the ATQ-N exhibits exceptional internal consistency reliability across clinical, inpatient, student, and community cohorts (Cronbach’s alpha ranging between α = .94 and .97; split-half reliability coefficients exceeding .90). Its temporal stability remains substantial among stable, non-depressed populations (test-retest correlations between r = .65 and .85), while demonstrating pronounced state-sensitivity following structured interventions such as Cognitive Behavioral Therapy (CBT) or pharmacotherapy. Furthermore, the questionnaire demonstrates robust convergent validity against standardized depressive severity metrics, including the Beck Depression Inventory (BDI) and the Hamilton Rating Scale for Depression (HRSD), while successfully discriminating depressed cohorts from normal controls and clinical populations exhibiting pure generalized anxiety disorders.

Keywords

Automatic Thoughts Questionnaire, ATQ-N, ATQ-30, negative automatic thoughts, cognitive triad, depression assessment, cognitive therapy, Steven D. Hollon, Philip C. Kendall, psychometrics, cognitive self-statements, depressive schemas, clinical validation, factor analysis, cognitive specificity

Authors

The Automatic Thoughts Questionnaire was developed and empirically validated through the collaborative academic partnership of two prominent clinical psychologists:

• Steven D. Hollon, Ph.D. — Gertrude Conaway Vanderbilt Professor of Psychology at Vanderbilt University (Nashville, Tennessee, USA). Dr. Hollon is an internationally celebrated authority in cognitive psychology, etiology, and empirical psychotherapeutic treatment of mood disorders, having served as a principal investigator on multiple landmark National Institute of Mental Health (NIMH) collaborative treatment studies.
• Philip C. Kendall, Ph.D., ABPP — Distinguished University Professor and Laura H. Carnell Professor of Psychology at Temple University (Philadelphia, Pennsylvania, USA). Dr. Kendall is a seminal figure in clinical child and adolescent psychology, cognitive-behavioral theory, and evidence-based interventions for internalizing disorders, widely known for his clinical research into cognitive restructuring and psychological assessment.

The initial construct development, pool compilation, item selection, and validation methodology were conducted within the Department of Psychology at the University of Minnesota and Temple University, culminating in the foundational 1980 monograph published in Cognitive Therapy and Research.

Purpose

The primary clinical and psychometric purpose of the Automatic Thoughts Questionnaire (ATQ-N) is the quantitative appraisal of the frequency, salience, and psychological endorsement of depressotypic automatic cognitions in individuals suffering from depressive syndromes, dysphoria, or affective spectrum disorders. In contrast to global symptom screening measures that conflate neurovegetative complaints (e.g., insomnia, appetite changes, somatic fatigue) with emotional disturbances, the ATQ-N was deliberately engineered to zero in specifically on the cognitive symptom cluster of unipolar depression.

Theoretical and Practical Rationale

Prior to the establishment of the ATQ-N in 1980, the measurement of cognitive distortions in depression relied primarily upon post-hoc clinical interviews, general mood self-reports, or unstructured thought-sampling methodologies (such as diary keeping or think-aloud protocols during stressful tasks). While clinically rich, these unstructured paradigms exhibited severe psychometric liabilities: low inter-rater reliability, prohibitive coding demands, social desirability distortions, and a failure to separate cognitive processing frequency from secondary distress. Hollon and Kendall recognized that to test the mechanistic claims of cognitive theory—namely, that changes in automatic cognition mediate recovery from depressive episodes—the discipline required an objective, standardized, easily administered psychometric scale with invariant content and validated scoring criteria.

Clinical Applications

In clinical practice, the ATQ-N serves multiple strategic functions across the assessment and intervention continuum:

• Baseline Cognitive Phenotyping: Administering the scale during clinical intake provides an objective profile of the patient’s dominant cognitive distortions, differentiating pervasive self-derogation from fatalistic future-oriented helplessness.
• Formulation of Behavioral Case Conceptualizations: Clinicians extract specific high-scoring items (e.g., Item 2: “I’m no good”, or Item 24: “I’ll never make it”) to populate cognitive conceptualization diagrams, identifying the cognitive targets for Socratic questioning, behavioral experiments, and decatastrophizing.
• Tracking Therapeutic Process and Treatment Response: Administered on a session-by-session or monthly basis, fluctuations in ATQ-N composite scores serve as an empirical proxy for the reduction of cognitive reactivity, reflecting whether cognitive interventions successfully suppress the activation of latent depressotypic schemas.
• Relapse Prevention and Vulnerability Profiling: Sustained cognitive reactivity on the ATQ-N following remission from major depressive episodes represents a demonstrated predictor of subsequent clinical recurrence, alerting clinicians to the necessity of booster sessions or maintenance cognitive therapy.

Research Applications

In psychopathology research and clinical trials, the ATQ-N represents an indispensable assessment tool:

• Treatment Mediation Research: Investigating whether cognitive therapy achieves its therapeutic outcomes specifically by dismantling negative self-referential cognitions, compared to pharmacotherapy (e.g., SSRIs, SNRIs) or behavioral activation alone.
• Cognitive Vulnerability and Diathesis-Stress Studies: Employing the scale during experimental mood-induction paradigms to observe cognitive priming effects in currently euthymic individuals with a history of recurrent depression.
• Cognitive Specificity Inquiries: Evaluating Beck’s cognitive specificity hypothesis by contrasting the cognitive profile of unipolar depression (characterized by themes of loss, personal failure, and abandonment captured on the ATQ-N) against the threat- and danger-oriented cognitions typical of anxiety disorders (measured via scales such as the Cognition Checklist or Anxious Thoughts Inventory).

Psychological Construct

The psychological construct evaluated by the ATQ-N is the phenomenon of Negative Automatic Thoughts (NATs). Within cognitive psychology, automatic thoughts are operationalized as discrete, situation-cued, involuntary, and rapid cognitive events that populate an individual’s stream of consciousness. Unlike reflective, deliberate problem-solving or analytical thinking, automatic thoughts emerge pre-reflectively, are accepted as self-evident truths without objective scrutiny, and are typically characterized by cognitive distortions (such as overgeneralization, dichotomous thinking, mental filtering, and personalization).

Dimensions and Subscales of the ATQ-N

Extensive factor-analytic evaluations across clinical and non-clinical samples have established that the 30 items of the ATQ-N coalesce around four distinct yet moderately intercorrelated first-order dimensions, each representing an aspect of depressotypic phenomenology:

1. Personal Maladjustment and Desire for Change (PMDC)

This subscale captures the acute, disquieting subjective awareness of cognitive-affective decompensation, accompanied by an intense yearning for intrapsychic transformation or psychological escape. It operationalizes the intrapsychic tension between one’s current state of perceived breakdown and the longing for normalcy. Representative items loading heavily on this factor include:

• Item 7: “I wish I were a better person.”
• Item 10: “I’m so disappointed in myself.”
• Item 14: “What’s wrong with me?”
• Item 20: “What’s the matter with me?”
• Item 26: “Something has to change.”

Elevated scores on PMDC reflect the ego-dystonic frustration of feeling alienated from one’s own baseline cognitive and emotional efficacy, serving as an index of subjective distress and perceived internal defectiveness.

2. Negative Self-Concepts and Negative Expectations (NSNE)

The NSNE dimension reflects global, characterological self-denigration paired with deterministic, pessimistic forecasts regarding personal capability and future trajectory. Grounded in Beck’s cognitive triad, this factor reflects deeply entrenched negative beliefs regarding the ongoing self and the future. Representative items include:

• Item 2: “I’m no good.”
• Item 3: “Why can’t I ever succeed?”
• Item 9: “My life’s not going the way I want it to.”
• Item 21: “I’m a loser.”
• Item 23: “I’m a failure.”
• Item 24: “I’ll never make it.”
• Item 28: “My future is bleak.”

These self-statements epitomize fatalistic cognitive patterns where transient setbacks are attributed to internal, stable, and global deficiencies, solidifying expectations of inexorable defeat.

3. Low Self-Esteem (LSE)

The Low Self-Esteem subscale measures raw, unmediated self-directed hostility and the affective sense of total personal invalidation. It captures profound moral and existential worthlessness, moving beyond dissatisfaction into affective self-repudiation. The primary marker items identifying this factor are:

• Item 17: “I hate myself.”
• Item 18: “I’m worthless.”

High endorsements on the LSE dimension represent a clinically significant marker of deep depressive severity and have been shown to correlate strongly with suicidal ideation, reflecting the collapse of self-compassion and perceived self-value.

4. Helplessness

The Helplessness dimension captures the perceived inability to initiate purposeful, goal-directed action or to derive value from continued persistence. It mirrors the behavioral and motivational deficits observed in learned helplessness paradigms, where agency is perceived as entirely decoupled from positive outcomes. The defining items on this dimension include:

• Item 29: “It’s just not worth it.”
• Item 30: “I can’t finish anything.”

Endorsement of these thoughts reflects severe motivational abulia, reinforcing behavioral avoidance, withdrawal, and clinical lethargy.

Theoretical Framework

The Automatic Thoughts Questionnaire is situated within Beck’s cognitive theory of depression (Beck et al., 1979). Beck posited that unipolar depression is characterized by a persistent structural disturbance in information processing. This conceptual framework is organized around three foundational constructs: the Cognitive Triad, Core Schemas, and Cognitive Errors.

Figure 1: Schematic representation of Beck’s Cognitive Triad illustrating negative appraisals of the Self, the World/Environment, and the Future.

1. The Cognitive Triad

At the center of Beck’s model is the negative cognitive triad: systematic, negative biases regarding:

• The Self: Depressed individuals perceive themselves as intrinsically defective, inadequate, diseased, or lacking the requisite resources to achieve happiness (e.g., ATQ Item 18: “I’m worthless”; Item 8: “I’m so weak”).
• The World / Immediate Experience: Environmental demands are perceived as overwhelming, excessively burdensome, and unfair, with the individual feeling unassisted and alienated (e.g., ATQ Item 1: “I feel like I’m up against the world”; Item 4: “No one understands me”).
• The Future: The individual anticipates that current suffering, deficits, and failures will persist indefinitely, breeding despair and resignation (e.g., ATQ Item 24: “I’ll never make it”; Item 28: “My future is bleak”).

2. Depressotypic Schemas and Latency

Cognitive theory postulates that early developmental adversity, rejection, or trauma precipitates the formation of latent cognitive structures termed depressive schemas or core beliefs (e.g., “I am unlovable,” “I am incompetent”). During euthymic periods, these schemas remain quiescent and dormant. However, when triggered by life events matching the schema’s thematic content (such as interpersonal rejection, academic failure, or occupational loss), these schemas become hyper-activated, dominating the individual’s executive attention and filtering mechanisms.

The ATQ-N does not measure schemas directly—which are tacit, deeply embedded, and often non-conscious structures—but rather measures their conscious cognitive manifestations: the explicit negative automatic thoughts that surface into working memory.

3. The Mediational Role of Automatic Cognitions

A central tenet of the cognitive-behavioral model is that thoughts act as proximate mediators of affect and behavior. An external triggering event does not directly generate depressive affect; rather, it is the subjective, biased cognitive appraisal of that event that fuels sadness, shame, hopelessness, and withdrawal. The ATQ-N was constructed to evaluate this mediational link. By tracking the decline of automatic thoughts during psychotherapy, researchers confirmed that cognitive restructuring reliably precedes and predicts the alleviation of depressive symptoms, validating the directional architecture of Beck’s model.

Validity

The psychometric validity of the ATQ-N has been examined across a variety of clinical, community, cross-cultural, and developmental samples.

Construct and Structural Validity

In the original validation study by Hollon and Kendall (1980), the scale was administered to a primary development sample of 312 undergraduate students and cross-validated on an independent sample of 78 psychiatric outpatients and non-depressed control subjects. Factor analyses confirmed that the 30 items formed a coherent construct with an eigenvalue-to-variance ratio validating a dominant primary dimension accounting for roughly 45% of the total variance, while simultaneously sustaining the four distinct first-order subscales described above.

Convergent Validity

The ATQ-N demonstrates convergent correlations with established criteria of depressive severity across multiple assessment modalities:

• Beck Depression Inventory (BDI): Correlations between the ATQ-N and BDI are consistently observed between r = .68 and r = .86 across both non-clinical student groups and diagnosed clinical psychiatric populations (Hollon & Kendall, 1980; Harrell & Ryon, 1983).
• Hamilton Rating Scale for Depression (HRSD): Clinician-rated depressive severity correlates significantly with ATQ-N scores (ranging from r = .55 to r = .70), demonstrating that self-reported automatic thoughts reflect objective clinical markers of the disorder.
• Dysfunctional Attitude Scale (DAS): ATQ-N scores correlate moderately to strongly (r = .50 to .65) with the DAS, which evaluates underlying depressotypic conditional assumptions, reflecting convergent theoretical alignment between latent schemas and explicit automatic thoughts.

Discriminant and Discriminative Validity

A primary psychometric requirement during the scale’s development was its capacity to differentiate clinically depressed patients from both healthy populations and non-depressed psychiatric patients:

• Depressed vs. Non-Depressed Samples: Hollon and Kendall (1980) demonstrated that clinically depressed patients yielded a mean score of 79.64 (SD = 22.29), compared to non-depressed control means of 48.57 (SD = 10.89), a statistically significant difference demonstrating large effect sizes (Cohen’s d > 1.5).
• Cognitive Specificity (Depression vs. Anxiety): In a critical investigation of cognitive content specificity, Hollon, Kendall, and Lumry (1986) evaluated ATQ-N profiles across unipolar depressed patients, mixed anxious-depressed patients, pure generalized anxiety disorder patients, and psychiatric controls. The ATQ-N distinguished depressed patients from pure anxious cohorts, corroborating that negative thoughts regarding personal loss, worthlessness, and defeat are specific to depression, whereas anxious cognitions center on threat and vulnerability.

Predictive and Treatment Sensitivity Validity

The ATQ-N is sensitive to change following targeted clinical interventions. In clinical trials comparing cognitive therapy, pharmacotherapy (tricyclic antidepressants or SSRIs), and combined treatments (Dobson et al., 1991), ATQ-N scores declined systematically alongside clinical improvement. Reductions in ATQ-N scores during early therapy sessions significantly predict post-treatment remission, establishing the predictive utility of the scale.

Reliability

The Automatic Thoughts Questionnaire exhibits excellent reliability across a broad range of psychometric indices, demonstrating both high measurement precision and situational stability.

Internal Consistency Reliability

Across diverse empirical investigations, the ATQ-N shows high internal consistency:

• In the original Hollon and Kendall (1980) validation investigations, the coefficient alpha was documented at an exceptional α = .97 for student cohorts and α = .96 in clinical outpatients.
• Split-half reliability calculations yielded an equivalent coefficient of r = .97, demonstrating homogeneous item-total relationships across the instrument.
• Subsequent clinical validation by Harrell and Ryon (1983) replicated these indices with psychiatric inpatients and outpatients, yielding Cronbach’s alpha values between α = .95 and α = .97.
• Shortened variants, such as the ATQ-15 and ATQ-8 validated by Netemeyer et al. (2002), maintain high internal consistency values ranging from α = .88 to α = .93.

Item-Total Correlations

Corrected item-total correlations across the 30 items are robust, with virtually all items exceeding r = .50, and the majority falling between r = .60 and r = .81. Items such as Item 2 (“I’m no good”), Item 18 (“I’m worthless”), and Item 23 (“I’m a failure”) consistently exhibit the highest discrimination indices, functioning as psychometric markers of the core underlying construct.

Test-Retest Stability

Because negative automatic thoughts are conceptualized as state-dependent cognitive events that fluctuate based on mood and contextual stressors, test-retest reliability varies naturally as a function of the clinical sample stability:

• In non-depressed, untreated control populations measured over 1- to 3-week intervals, temporal stability coefficients remain high, typically ranging between r = .65 and r = .85.
• In clinical cohorts undergoing active psychotherapy or pharmacotherapy, test-retest correlations decrease appropriately over longer observation windows (e.g., 8 to 12 weeks), reflecting the instrument’s clinical sensitivity to therapeutic improvement rather than measurement unreliability.

Factor Analysis

The latent structural architecture of the ATQ-N has been systematically evaluated through both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) across clinical, student, and cross-cultural cohorts.

Original Exploratory Factor Analysis (Hollon & Kendall, 1980)

In the seminal psychometric derivation, Hollon and Kendall subjected the initial 45 prospective items to principal components analysis followed by orthogonal (Varimax) and oblique (Promax) rotations using an unselected undergraduate sample (N = 312). Using Cattell’s scree test criteria and eigenvalue retention thresholds (> 1.0), the researchers retained 30 items loading cleanly onto four distinct primary factors. Items were retained if they met a factor loading criterion of ≥ .40 on a primary factor with minimal secondary cross-loadings:

While the four factors delineate identifiable phenomenological themes, the moderate-to-high intercorrelations among the factors (ranging from r = .52 to .74) confirmed that a hierarchical, higher-order general factor—representing overall General Negative Cognitive Bias—subsumes these specific domains.

Confirmatory Factor Analysis and Structural Refinement

Later investigations utilizing CFA have assessed whether the ATQ-30 is best captured via a single-factor, a four-factor correlated model, or a higher-order hierarchical model:

• Netemeyer et al. (2002): Evaluated the full 30-item scale alongside shortened forms. Across both non-clinical and psychiatric cohorts, the hierarchical model demonstrating four first-order factors loading onto a single higher-order Negative Cognition dimension demonstrated acceptable fit (CFI > .92, RMSEA ≈ .065).
• Cross-Cultural CFA Invariance: International psychometric evaluations (e.g., Persian version by Ghassemzadeh et al., 2006; Turkish, Spanish, and Chinese adaptations) have demonstrated stable structural validity, confirming that the factor architecture remains invariant across languages and cultural settings.

Instrument / Measurement Tool

The structured technical parameters, item count, and administrative scoring guidelines for the Automatic Thoughts Questionnaire are outlined below:

• Instrument Type: Standardized self-report psychometric rating scale.
• Format: Pen-and-paper, computer-assisted psychological test, or secure online assessment battery.
• Item Count: Exactly 30 negative self-referential statements. Shortened psychometric variations include the 15-item version (ATQ-15: items 2, 6, 10, 11, 12, 13, 14, 18, 24, 25, 26, 27, 28, 29, 30) and the 8-item version (ATQ-8: items 2, 10, 14, 18, 25, 26, 28, 30).
• Recall Period: Standard instructions anchor assessment to the past week, including today.
• Response Dimensions:
  • Primary Dimension (Frequency): 5-point Likert scale:
    • 1 = Not at all (or Never)
    • 2 = Sometimes (or Rarely / Once in a while)
    • 3 = Moderately (or Fairly often)
    • 4 = Often (or Very often)
    • 5 = All the time (or Always)
  • Secondary Dimension (Optional Clinical Rating – Degree of Belief): 5-point Likert scale:
    • 1 = Not at all
    • 2 = Somewhat
    • 3 = Moderately
    • 4 = Very much
    • 5 = Totally
• Scoring Rules:
  • All 30 items are positively keyed in the direction of depressive cognitions; there are no reverse-scored items.
  • Frequency Total Score: Calculated by summing the numerical ratings (1 through 5) across all 30 items. Theoretical range spans from 30 to 150.
  • Belief Total Score: When administered, calculated by summing belief ratings (1 through 5) across all 30 items (range: 30 to 150).
  • Subscale Scores: Derived by summing item subsets:
    • Personal Maladjustment and Desire for Change (PMDC): Items 7, 10, 14, 20, 26 (Score range: 5 to 25).
    • Negative Self-Concepts and Negative Expectations (NSNE): Items 2, 3, 9, 21, 23, 24, 28 (Score range: 7 to 35).
    • Low Self-Esteem (LSE): Items 17, 18 (Score range: 2 to 10).
    • Helplessness: Items 29, 30 (Score range: 2 to 10).
• Clinical Cutoff Interpretations (General Normative Guidelines):
  • 30 – 49: Normal, non-depressed range; minimal depressotypic thinking.
  • 50 – 65: Mild cognitive dysphoria; sporadic negative self-statements.
  • 66 – 85: Moderate automatic negative cognition; indicative of clinically relevant depressive distress.
  • 86 – 150: Severe, pervasive cognitive entrenchment typical of major unipolar depression.
• Administration Time: Approximately 5 to 10 minutes for completion.

Permissions & Fee and Test Year

The Automatic Thoughts Questionnaire was developed by Steven D. Hollon and Philip C. Kendall and officially introduced in the scientific literature in 1980. The copyright for the initial publication resides with the original publisher, Springer Science+Business Media (formerly Plenum Publishing Corporation) via the journal Cognitive Therapy and Research.

For non-commercial academic research, pedagogical purposes, and standard clinical practice, the authors and clinical traditions have historically permitted broad access to the scale without licensing fees, provided that standard academic attribution and citations are maintained. Commercial distribution, translation into new proprietary digital systems, incorporation into fee-bearing electronic health record systems, or inclusion in sponsored clinical trial outcome batteries generally requires formal copyright clearance or commercial licensing approval from the copyright holders or authors.

References

The academic validation and clinical implementation of the ATQ-N are substantiated by the following peer-reviewed publications:

• Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression. Guilford Press.
• Dobson, K. S., Hollon, S. D., & Evans, M. D. (1991). Cognitive therapy and pharmacotherapy for depression: So which is better? Clinical Psychology Review, 11(4), 449–479. https://doi.org/10.1016/0272-7358(91)90117-G
• Ghassemzadeh, H., Mojtabai, R., Karamghadiri, N., & Ebrahimkhani, N. (2006). Psychometric properties of a Persian-language version of the Automatic Thoughts Questionnaire: ATQ-Persian. International Journal of Social Psychiatry, 52(2), 127–137. https://doi.org/10.1177/0020764006062095
• Harrell, T. H., & Ryon, N. B. (1983). Cognitive-behavioral assessment of depression: Clinical validation of the Automatic Thoughts Questionnaire. Journal of Consulting and Clinical Psychology, 51(5), 721–725. https://doi.org/10.1037/0022-006X.51.5.721
• Hollon, S. D., & Kendall, P. C. (1980). Cognitive self-statements in depression: Development of an Automatic Thoughts Questionnaire. Cognitive Therapy and Research, 4(4), 383–395. https://doi.org/10.1007/BF01173648
• Hollon, S. D., Kendall, P. C., & Lumry, A. (1986). Specificity of depressotypic cognitions in clinical depression. Journal of Abnormal Psychology, 95(1), 52–59. https://doi.org/10.1037/0021-843X.95.1.52
• Ingram, R. E., & Wisnicki, K. S. (1988). Assessment of positive automatic cognition. Journal of Consulting and Clinical Psychology, 56(6), 898–902. https://doi.org/10.1037/0022-006X.56.6.898
• Kazdin, A. E. (1990). Evaluation of the Automatic Thoughts Questionnaire: Negative cognitive processes and depression among children. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 2(1), 73–77. https://doi.org/10.1037/1040-3590.2.1.73
• Netemeyer, R. G., Williamson, D. A., Burton, S., Biswas, D., Jindal, S., Linda, S., & Netemeyer, D. A. (2002). Psychometric properties of shortened versions of the Automatic Thoughts Questionnaire. Educational and Psychological Measurement, 62(1), 111–129. https://doi.org/10.1177/0013164402062001008

Items of the Scale